Article Figures & Data
Figures
- Figure 1.
Pseudotyped recombinant rabies virus infects tectal neurons in the Xenopus tadpole. A, Schematic of the labeling strategies using recombinant SAD B19 rabies virus which has the glycoprotein deleted and replaced by EGFP (SADΔG-EGFP), rendering it incapable of transneuronal spread. Infection with B19G phenotypically complemented virus (top) relies on endogenous expression of the B19G receptor. Infection with EnvA pseudotyped virus (bottom) requires exogenous expression of its receptor, TVA, before viral injection. Co-labeling TVA-transfected neurons with tRFP allows them to be identified. Viral injections were made in the optic tectum, which is marked by a dashed box in the drawing to the left. B, SADΔG-EGFP(B19G) virus infects tectal neurons. Confocal Z-projection collected in vivo through the injected optic tectal lobe shows widespread virally-mediated expression of EGFP in infected neurons. C, SADΔG-EGFP(EnvA) virus infects tectal neurons transfected with TVA. The right optic tectal lobe was transfected with CMV::TVA/tRFP by whole-brain electroporation and injected with SADΔG-EGFP(EnvA) virus 4 d later. Confocal Z-projections collected in vivo through the optic tectal lobe electroporated with CMV::TVA/tRFP (magenta) and injected with SADΔG-EGFP(EnvA) virus (green). Neurons which co-express TVA/tRFP and viral EGFP are marked by yellow arrows. The remaining EGFP-expressing neurons lack detectable tRFP expression and are presumably invisible TVA-expressing neurons. D, Viral infection efficiency varies with developmental stage. Tadpoles at stages 42–43 (n = 16 tadpoles), 44–45 (n = 28 tadpoles), or 46–48 (n = 65 tadpoles) were electroporated with CMV::TVA/tRFP using whole-brain electroporation. Four days later, the transfected tectal lobe was injected with SADΔG-EGFP(EnvA) virus. The percentage of animals with EGFP-expressing neurons was highest between stages 44 and 48. E, Infection with SADΔG-EGFP(EnvA) virus requires TVA. Confocal Z-projection collected in vivo through an optic tectal lobe injected with SADΔG-EGFP(EnvA) shows no infected neurons in the absence of TVA electroporation. F, SADΔG-EGFP(EnvA) infects tectal neurons transfected with TVA driven by the VGAT promoter. The right optic tectal lobe was transfected with VGAT::gal4, UAS::TVA, and UAS::tRFP by whole-brain electroporation and injected with SADΔG-EGFP(EnvA) virus 4 d later. Confocal Z-projections collected in vivo through the optic tectal lobe showing electroporated (magenta) and infected (green) tectal neurons. Neurons which co-express TVA/tRFP and viral EGFP are marked by yellow arrows. The remaining EGFP-expressing neurons lack detectable tRFP expression and are invisible TVA-expressing neurons. G, Quantification of the proportion of invisible TVA cells per animal with and without amplification. Amplifying tRFP expression using the gal4-UAS system decreases the proportion of infected, EGFP+ cells that lack detectable tRFP compared with tRFP driven by the CMV promoter without amplification. Data are presented as mean ± SEM overlaid with individual data points (***p < 0.0001, Mann–Whitney test). H, Targeted electroporation of TVA/tRFP does not eliminate invisible TVA-expressing neurons. Micropipette-mediated electroporation was used to limit transfection with TVA/tRFP to one or few neurons in the right optic tectal lobe. Four days later, the electroporated tectal lobe was injected with SADΔG-EGFP(EnvA) virus. Confocal Z-projection collected in vivo through the optic tectal lobe shows that EGFP-expressing infected neurons which lack detectable tRFP are still present. Scale bars: 50 μm.
- Figure 2.
Transcomplementation with rabies glycoprotein does not result in transneuronal spread of recombinant rabies in tadpoles. A, Schematic of the monosynaptic tracing strategy using SADΔG-EGFP(EnvA) virus with transcomplementation of rabies glycoprotein, B19G. Neurons co-transfected with TVA/tRFP and B19G can be directly infected by EnvA pseudotyped virus through the TVA receptor. Viral particles which bud from directly infected neurons will have B19G on their surface because B19G is provided in trans. In mammals, those viral particles can infect presynaptic neurons through the endogenous B19G receptor. Because presynaptically infected neurons lack B19G expression, viral particles generated in those neurons lack the glycoprotein and are not infectious, thereby prohibiting further spread. B, In vivo time-lapse imaging of infected tectal neurons from 3–6 d following injection of SADΔG-EGFP(EnvA) virus in the presence or absence of B19G. One tectal lobe was transfected with TVA/tRFP (magenta) alone (left) or with TVA/tRFP and B19G (right) and then injected with SADΔG-EGFP(EnvA) virus 4 d later. At 3 and 6 d after viral injection, confocal Z-stacks through the tectal lobe were collected. Z-projections show an increase in the number of EGFP+ neurons without detectable tRFP (white arrows) from 3 to 6 d after viral injection in both the presence and absence of B19G. Scale bar: 50 μm. C, Quantification of the average fold-change in the number of EGFP-only cells from 3 to 6 d after injection. There is a similar increase in the average number of EGFP-only cells over time in the presence (n = 17 tadpoles) and absence (n = 7 tadpoles) of B19G, suggesting a lack of local presynaptic spread of rabies virus. Data are presented as mean ± SEM overlaid with individual data points (p = 0.74, Mann–Whitney U test).
- Figure 3.
Weak expression of B19G in vivo may explain the lack of transneuronal spread of rabies virus. A, B, Rabies glycoprotein is detected in the membrane fraction of transfected mammalian and Xenopus cell cultures by Western blotting. 293T (A) and XLK-WG Xenopus kidney cells (B) were transfected with B19G and proteins were extracted 24 h later. Membrane fractions were probed for B19G expression with anti-rabies glycoprotein antibody and β-tubulin acted as a loading control. Compared with untransfected cells, specific bands of ∼70 kDa were visible in transfected cells. Specific band in transfected XLK-WG cells is denoted by an arrow (B). C, Rabies glycoprotein is detected on the surface of Xenopus cells in vitro by immunocytochemistry. XLK-WG cells were transfected with GFP alone (top) or B19G and GFP (bottom). Confocal Z-projections of cells transfected with both B19G and GFP show surface expression of B19G by anti-rabies glycoprotein immunocytochemistry (magenta) without permeabilization. In contrast, no anti-rabies glycoprotein immunoreactivity is observed in cells transfected with GFP alone. Scale bar: 20 μm. D, Expression of B19G is very weak in vivo in tectal neurons. Tectal neurons were electroporated with CMV::B19G/tRFP, fixed 3–4 d later, and then immunohistochemistry with anti-rabies glycoprotein was performed. Confocal Z-projection of a 40-μm tissue slice shows very weak immunoreactivity for rabies glycoprotein (green) in B19G/tRFP expressing neurons (magenta). Scale bar: 50 μm.
- Figure 4.
Retrograde neuronal tracing using recombinant rabies virus. A, SADΔG-EGFP(B19G) virus retrogradely labels afferents to the injected target region. A montage of confocal Z-projections collected in vivo shows neurons infected by injection of SADΔG-EGFP(B19G) virus into the right tectal lobe. In addition to a large number of neurons expressing EGFP in the injected tectal lobe, retrogradely infected projection neurons are visible in the contralateral tectum, pretectum, and hindbrain. Tectal lobes are marked with dashed lines. Scale bar: 50 μm. B, A schematic which maps neurons labeled by unilateral tectal injection of SADΔG-EGFP(B19G) virus. Neurons in several regions known to project to the optic tectum are labeled. We also observed a large number of neurons in the injected tectal lobe and ipsilateral pretectum (green shading). AC, anterior commissure; FB, forebrain; HB, hindbrain; OT, optic tectum; PT, pretectum; V, ventricle. C, Retrograde viral tracing paired with immunohistochemistry reveals the cell types which project between the two tectal lobes. A montage of confocal Z-projections collected in vivo following injection of SADΔG-EGFP(B19G) virus into the right tectal lobe shows retrograde tracing of one intertectal neuron (boxed in left tectal lobe; i). Following fixation and tissue sectioning, immunohistochemistry was performed with an anti-GABA antibody to label inhibitory neurons. The EGFP+ neuron imaged in vivo (i) could be identified in fixed tissue slice (ii) and was GABA-negative (iii, magenta), suggesting that it is excitatory. Scale bars: 50 μm (i, ii) and 25 μm (iii). D, Schematic showing retrograde tracing strategy used in E. The left tectal lobe is electroporated with TVA/tRFP and 4 d later, SADΔG-EGFP(EnvA) virus is injected into the right tectal lobe. Expression of TVA on the surface of intertectal axons mediates viral infection of intertectal neurons in the left hemisphere. E, SADΔG-EGFP(EnvA) virus can be used to retrogradely trace neurons defined by anatomic location and axonal projections using promoter-driven expression of TVA. A montage of confocal Z-projections collected in vivo (i) demonstrate retrograde tracing of TVA-expressing neurons in the left tectal lobe (magenta) following injection of SADΔG-EGFP(EnvA) virus into the right tectal lobe. Retrogradely infected neurons (boxed in i) are shown at higher magnification (ii) and cells co-expressing EGFP and tRFP are marked by yellow arrows. Scale bars: 50 μm (i) and 25 μm (ii).
Tables
Data structure Type of test p value Sample size (n = tadpoles) a Fisher’s exact test 0.37 22°C n = 12
28°C n = 11b Fisher’s exact test 0.63 CMV::gal4 n = 4
VGAT::gal4 n = 21c Not normally distributed Mann–Whitney U test <0.0001 gal4-UAS amplification n = 9
No gal4-UAS amplification n = 14d Not normally distributed Mann–Whitney U test 0.74 B19G+ n = 17
B19G– n = 11For each statistical test run in the study, the data structure, statistical test, p value, and sample size are listed.
In this issue
